摘要
针对常用遗传算法存在容易产生过早收敛的问题,提出了一种将强制变异、最佳解保留和自适应交叉变异参数调整相结合的改进遗传算法。这种方法将进化过程中群体的平均适应度与最大适应度进行比较,以确定是否需要对群体实施强制变异或采用自适应交叉、变异概率调整。数值模拟的结果表明,这种方法可有效地克服早熟现象。
An improved genetic algorithm is developed, with the addition of three new operations: enforced mutation; direct preserving of best chromosome and using adaptive parameters. The method compares average fitness of pop with maximum fitness, and makes sure the necessity of enforced mutation or crossover and mutation with adaptive parameters. The simulating results indicate the improved method can prevent premature and realize global optimization effectively.
出处
《中国科学院研究生院学报》
CAS
CSCD
2003年第3期316-320,共5页
Journal of the Graduate School of the Chinese Academy of Sciences
基金
国家自然科学基金资助项目(2 0 0 73 0 42)
关键词
遗传算法
早熟
强制变异
自适应参数
genetic algorithm, premature, enforced mutation, adaptive parameters